Article
Computer Science, Information Systems
Yongqi Li, Nan Yang, Liang Wang, Furu Wei, Wenjie Li
Summary: To address the issue of question ambiguity in conversation question answering, a generative retrieval method called GCoQA is proposed. GCoQA retrieves passages by generating their identifiers token-by-token via an encoder-decoder architecture, leading to significant improvements in passage retrieval and document retrieval compared to current methods.
INFORMATION PROCESSING & MANAGEMENT
(2023)
Article
Computer Science, Artificial Intelligence
Ivo Pisacovic, Frantisek Darena, David Prochazka, Vit Janis
Summary: Establishing normative documents is essential for larger organizations to control processes and provide solutions to common problems, but the formal and difficult-to-read nature of these documents necessitates different customer services. Companies are increasingly developing chatbots for firstline customer support automation, but automatic answering directly from normative documents is often ineffective. A novel preprocessing method is proposed in this paper to improve the accuracy of automatic question answering on normative documents.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Maxwell A. Weinzierl, Sanda M. Harabagiu
Summary: This study designed a QA system capable of answering ad-hoc questions about COVID-19, using automatic question generation and question entailment. The results showed that the system achieved state-of-the-art performance with expert questions and competitive performance with consumer questions. However, more than half of the answers were missed due to limitations in the relevance models used. Improvements should be considered for better relevance models and enhanced inference methods.
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
(2023)
Article
Computer Science, Artificial Intelligence
Munazza Zaib, Wei Emma Zhang, Quan Z. Sheng, Adnan Mahmood, Yang Zhang
Summary: This article discusses the development of question answering systems, with a particular focus on the importance and current status of multi-turn question answering. It provides a comprehensive review of recent research based on reviewed papers. The findings highlight the significance of multi-turn question answering in advancing the field of conversational artificial intelligence.
KNOWLEDGE AND INFORMATION SYSTEMS
(2022)
Article
Multidisciplinary Sciences
Sandeep Varma, Shivam Shivam, Snigdha Biswas, Pritam Saha, Khushi Jalan
Summary: This paper utilizes conversational analytics tool and Knowledge Graph for querying in the domain of tabular data, avoiding traditional semantic parsing approach and providing accurate answers to different types of queries.
Article
Computer Science, Information Systems
Huiyong Wang, Ding Yang, Liang Guo, Xiaoming Zhang
Summary: This study aims to build a joint task model with generalization ability and benchmark its performance. A deep-learning-based approach was used for intent detection and slot filling, with improvements to the LSTM network structure. The proposed model outperformed other benchmark methods, especially on the computer science literature question dataset.
DATA TECHNOLOGIES AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Yongrui Chen, Huiying Li, Guilin Qi, Tianxing Wu, Tenggou Wang
Summary: Query graph construction aims to generate a correct SPARQL query to answer natural language questions on a knowledge graph. Existing methods face challenges in handling complex questions including complicated SPARQL syntax, huge search space, and locally ambiguous query graphs. This paper proposes a novel end-to-end approach that leverages a unified graph grammar called AQG and hierarchical autoregressive decoding to construct query graphs effectively. Experimental results demonstrate that the proposed method significantly improves the state-of-the-art performance on complex KGQA benchmarks.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Jinting Lu, Xiaobing Sun, Bin Li, Lili Bo, Tao Zhang
Summary: Software bugs are common in the software development process. To fix bugs quickly, developers need to understand bug information efficiently. Existing approaches often fail to fully express users' intents, resulting in irrelevant results. BEAT utilizes templates to understand natural language questions and generate structured queries, achieving better F1-score values than existing Q&A approaches.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Information Systems
Yi-Hui Chen, Eric Jui-Lin Lu, Ying-Yen Lin
Summary: The study introduces a low-cost SPARQL generator called Light-QAWizard, which utilizes multi-label classification to reduce query cost significantly. By integrating the results of a template classifier, Light-QAWizard generates corresponding query grammars that outperformed all other models with nearly half the query cost.
Article
Computer Science, Artificial Intelligence
Tahani H. Alwaneen, Aqil M. Azmi, Hatim A. Aboalsamh, Erik Cambria, Amir Hussain
Summary: Question answering is a subfield of information retrieval that aims to answer questions posed in natural language. The development of Arabic question answering systems has been hindered by linguistic challenges and limited resources. Research in this area includes examining challenges, existing systems, techniques, and future directions for Arabic question answering systems.
ARTIFICIAL INTELLIGENCE REVIEW
(2022)
Article
Computer Science, Hardware & Architecture
Jun-Feng Fan, Mei-Ling Wang, Chang-Liang Li, Zi-Qiang Zhu, Lu Mao
Summary: This paper proposes a new framework for joint intent prediction and slot filling, explicitly establishing the correlation between intent and slots, and improving semantic results by introducing slot recognition information and slot-gated mechanism.
JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY
(2022)
Article
Computer Science, Information Systems
Yongqi Li, Wenjie Li, Liqiang Nie
Summary: In this article, the authors discuss the challenges of building an effective conversational open-domain QA system and propose an end-to-end Dynamic Graph Reasoning approach called DGRCoQA. The proposed approach demonstrates significant improvements over existing methods in experiments.
ACM TRANSACTIONS ON INFORMATION SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Shivansh Gupta, Sanju Tiwari, Fernando Ortiz-Rodriguez, Ronak Panchal
Summary: The research introduces a knowledge graph-based natural language query-answering model designed for Indian Missiles, featuring a Missile Knowledge Graph with 177 entities linked by 400 relationships. The study explores its application in parsing complex natural language questions and generating appropriate answers.
Article
Computer Science, Artificial Intelligence
Yang Deng, Wenxuan Zhang, Weiwen Xu, Ying Shen, Wai Lam
Summary: Nonfactoid question answering (QA) is a challenging application in natural language processing (NLP). This work proposes a graph-enhanced multihop query-focused summarizer (GMQS) to handle nonfactoid QA. The method utilizes graph-enhanced reasoning techniques to capture semantic relations and uses a relational graph attention network (RGAT) for information aggregation. Experimental results show that GMQS outperforms existing methods on two nonfactoid QA datasets and enables explainable multihop reasoning.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Automation & Control Systems
Hongyin Zhu, Prayag Tiwari, Ahmed Ghoneim, M. Shamim Hossain
Summary: Effective models are needed in the AIoT field to understand and answer questions. This article proposes RoBERTa(AIoT) to address the lack of high-quality labeled AIoT OA datasets. Pretraining on an AIoT corpus improves the model's performance on AIoT OA tasks. Experimental results show significant improvements.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Multidisciplinary Sciences
Hsiu-Min Chuang, Fanpyn Liu, Chung-Hsien Tsai
Summary: The study investigates the feasibility of using artificial intelligence technology to detect abnormal attacks in SDN networks. By applying machine learning algorithms and a hierarchical multi-class architecture, the study successfully achieves early detection of DDoS attacks and improves the performance of minority classes. Experimental results show that decision tree, random forest, bagging, AdaBoost, and deep learning models exhibit the best performance in this regard.